New Edges First Method

I am not sure if this has been thought of before, so I will put it down.

The basic idea of this method is to solve the edges first, then the corners. I have only just thought of this, like in the past 30 minutes, so no algorithms.

Basic outline:

1.) Make cross on bottom. 0 Algs
2.) Solve edges in E Slice. 0 Algs
3.) Orient last 4 edges. 3 Algs
4.) Put all corners for one face on that face, but not in correct orientation/permutation. By doing this, the same thing will happen to the opposite face. 6 Algs (Not positive)
5.) Use COLL algs on one side. 40 Algs.
6.) Permute Edges on same face. 4 algs.
7.) Repeat steps 5+6 on opposite face.

Total algs needed: 53

Can anyone suggest a prgram to generate the algorithms for step 4? I am not able to find one.

It would work, but it would be very slow. For the E layer edges, it would just be faster to bring along the corner too. That would leave you with just the last layer. Then you are essentially doing Fridrich.

All edges: 0 algs (see Heise or Garron).
Intuitive for "triple excross with optional solving of corners, then intuitively solve one LL edge at a time.

All corners: 0 algs (see Heise, Hardwick, Beyer, or Ville).
Commutators.

Test solve: Edges in 20 moves, Corners in 55, solving only 1 corner at a time because I suck at using commutators without insertions. Ville can probably do it in less than 30 moves.

It is possible to do all the edges and corners intuitively, but that would take a lot of moves and more time than using an algorithm.